Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Advertise on LowEndTalk.com
Virtualizor Bugs
New on LowEndTalk? Please read our 'Community Rules' by clicking on it in the right menu!

Virtualizor Bugs

NexHostNexHost Member
edited August 2015 in Reviews

Hi,

I'm posting my genuine worries and concerns about Virtualizor.

I switched from SolusVM earlier on this year to Virtualizor. and since then the Product has been nothing but a pain in the ass. And I'm not saying it to cause drama it's true.

Try having a Product that is poorly documented. or a product that has features but with no clear indication that if you have it setup a specific way. Your VMS will loose their data upon them being migrated across another node again not enough documentation. Not straight away this happens but months later when a client reboots their VPS. this is caused by the fact that if you use file based storage. their code does not support file based migrations.

My second major concern is that they release code that is buggy. their Multivirt Feature is completely useless. Random shutdown(s) occur. and all Virtualizor can do is blame it on a Hardware Issue?

 2015-08-05 09:39:25.051+0000: 14820: error : qemuRemoveCgroup:781 : internal error         Unable to find cgroup for v1002

 2015-08-05 09:39:25.051+0000: 14820: error : qemuAutostartDomain:294 : Failed to autostart VM 'v1002': Unable to create cgroup for v1002: No such file or directory
 2015-08-05 09:39:25.066+0000: 14820: error : virFileReadAll:462 : Failed to open file   '/vz/root/743/sys/fs/cgroup/memory/libvirt/qemu/v1003/memory.use_hierarchy': No such file or directory

I allowed Virtualizor tech access to this machine weeks after originally reporting the error to them. All I got from them is that the issue is fixed please check. So I did. Everytime I install a new KVM VPS on this 1 specific node. it will shutdown the VM. How ever Yesterday I contact ed virtualizor again because another server where a couple of the VPS have been put during the past several weeks. is now not coping with the Load.

 qemu: terminating on signal 15 from pid 822047

This error above appears everytime the VPS is shutdown. how ever I have over 7 + Licenses with Virtualizor. any of the Nodes not running Multivirt keep having the same exact error above unable to start VPS due to a cgroup error.

Oh and I forgot to mention if you have VPS running from a ISO if you ever use the stop/start function it can screw up the disk partitioning. Something I have never had a problem with. All I got from them again is backup your critical VMs ?

«13

Comments

  • WebProjectWebProject Member, Provider

    jmckeag12 said: any good alternatives will be considered

    highly recommended to use SolusVM, therefore you need to switch back to stable control panel.

    VPS Price Match Guarantee on: All our range of DDOS protected XEN-HVM VPS Plans
    Are you looking for best price for self-managed VPS? See WebProVPS website for more details.
  • NexHostNexHost Member
    edited August 2015

    @rahulks said:
    virtualizor

    Yes include them so they know their product is complete and utter shit.

    Oh another bug I have reported to them with no fix is their stupid DNS System. every time I modify a record it screws up the main DNS Record. and they could not reproduce it either.

    The record ends up just looking like the following.

     2015080300 28800 7200 604800 86400
    

    When it should be the following

     namserver.com [email protected] 2003080800 28800 7200 1209600 60
    
  • WebProject said: therefore you need to switch back to stable control panel.

    LOL

    I AM BACK :)
    Working Windows Server 2012 R2 on 6GB! Beat that!

  • AnthonySmithAnthonySmith Top Provider

    For once we agree on something @jmckeag12 I would not worry about it though, no doubt solusvm v2 which is imminent will have an importer to get you back on track.

    Thanked by 1NexHost
  • @jmckeag12

    I have to agree, in the past I was fairly + with Virtualizor, however they just seems to be focusing on new wonderful features and not fixing the core system & bugs.

    Lets hope SolusVM2 does have support to migrate into from Virtualizor.

  • joepie91joepie91 Member, Provider

    jmckeag12 said: All I got from them is that the issue is fixed please check.

    That sounds like outsourced support.

  • NexHostNexHost Member
    edited August 2015

    @AnthonySmith said:
    For once we agree on something jmckeag12 I would not worry about it though, no doubt solusvm v2 which is imminent will have an importer to get you back on track.

    I hope SolusVM will have a Importer and support for Operating Systems other then RHEL. moved from SolusVM because after so long of a box being up. weird stuff would start happening. Such as modprobe going in a loop with 3,000 + processes. until eventually the box would die.

    @AshleyUk said:
    jmckeag12

    I have to agree, in the past I was fairly + with Virtualizor, however they just seems to be focusing on new wonderful features and not fixing the core system & bugs.

    Lets hope SolusVM2 does have support to migrate into from Virtualizor.

    Yes and I have a node that is hitting more then 50% Load. having to accommodate other VPS because the main machine is not working.

    http://i.gyazo.com/2373960ab17ddb5e4d4af0a6e2ff9389.png

    joepie91 said: That sounds like outsourced support.

    Seems like they are using outsourced support. same as SolusVM.

  • LicensecartLicensecart Member
    edited August 2015

    @AshleyUk said:
    jmckeag12

    I have to agree, in the past I was fairly + with Virtualizor, however they just seems to be focusing on new wonderful features and not fixing the core system & bugs.

    Lets hope SolusVM2 does have support to migrate into from Virtualizor.

    There is an importer on SolusVM2 :) I can promise you that.

    Blesta.Store - Blesta specialists
    Atlanical - Social Network Management

  • NexHostNexHost Member
    edited August 2015

    Another day nearly over and still no fix.

  • Virtualizor seems to be a love or hate type thing!

  • even thought I am apparently a "fucktard", @jmckeag12 I'm happy to try to recreate this problem on my servers

    Have I had problems with Virtualizor? yes! But their support team have been very helpful, and quick to solve problems.

    if I can recreate the same problems, then that helps Virtualizor to know it's not just your servers, and hopefully resolve your issues sooner. it also in my interests to check on this, so I don't fall into the same hole.

    free trial zilore monitoring

  • NexHostNexHost Member
    edited August 2015

    Bruce said: even thought I am apparently a "fucktard", @jmckeag12 I'm happy to try to recreate this problem on my servers

    You called me a lazy bastard what do you expect?

    I have had other people reproduce the same errors. the support have verified that there is indeed a issue going on. And Virtualizor have contacted me via PM who are putting me in touch with the lead dev. I have also reproduced the same error on my own machines.

    Also this is not my server Bruce. it's a client machine..

  • I'd say that Softaculous have been EXTREMELY responsive to our requests, same day or better consistently.

    If you have accurately documented the bug in a way it can be reproduced by them, then I'd expect them to solve your issue within a very short time.

  • VereloxVerelox Member, Provider

    MarkTurner said: If you have accurately documented the bug in a way it can be reproduced by them, then I'd expect them to solve your issue within a very short time.

    I hate to disagree with you but during our experience, every time we would open a ticket to ask for assistance or report a bug, they would usually first of all ask us for the root password which we refuse to provide (that we started explaining our refusal for providing the root password in future tickets), then we would have to wait a few hours/days for a constructive response from them because their previous responses would either ask us for steps to reproduce the issue (although we attach that in the first post if applicable), inform us that they are checking with the developers or ask us for the root password again then finally attempt to fix the issue. The control panel was also full of bugs involving migration, server suspension/termination (where it would terminate/suspend the wrong machine, but they were able to fix this after 2 days of exchanging messages on their ticketing system - this caused us a lot of issues as we would eventually attempt to recover accidentally deleted LVMs), bandwidth calculation, issues with IPv6, automatic selection of servers based on server groups (they also had an issue regarding this with the WHMCS module) and I believe a few more bugs we talked with them about.

    I think however that they made a lot of changes and improvements since the last time we were using their control panel hence I don't like to say that I don't recommend their control panel for production use, but we really wish they would pay more attention to the bugs raised by users and try to eliminate some of the major bugs that makes the control panel unstable as their support was one of the major reasons we decided to stop using their services.

    Thanked by 1NexHost
  • agentmishraagentmishra Member, Provider

    so virtualizor has become like hypervm...

    skype id : agentmishra, gtalk id : agentmishra, email : [email protected]

  • Verelox said: I hate to disagree with you but during our experience, every time we would open a ticket to ask for assistance or report a bug, they would usually first of all ask us for the root password which we refuse to provide

    I think everyone's experience will be different, I know for us they've bent over backwards to assist. The software is far from bug free, but when we make reports we get fast results. At the end of the day thats all we can ask, we can expect 100% bug free software.

    I agree their focus should be on bug fixes for a few months rather than rolling out features. I know that customers of ours who have experienced issues and reported them back via us, we have been able to get instant attention for them.

    agentmishra said: so virtualizor has become like hypervm...

    Nothing like that, I think these guys are focused on their product, they need to refocus onto bug fixes for a while but they're definitely going in the right direction.

  • the documentation is far from acceptable. if they improved the docs they'd have less tickets

    there's too many issues with stuff not working unless you do it exactly as it was designed to be used. often that isn't obvious, so you fall down a hole. when stuff doesn't work, give the user a clue it didn't work. there's little debug, especially in the WHMCS module. again, if they'd provided better feedback, they'd get fewer tickets

    free trial zilore monitoring

  • Bruce said: especially in the WHMCS module

    You need to write your own, their WHMCS and Hostbill modules are crap.

  • @MarkTurner said:

    that might be the best solution, but doesn't fit well with a product you pay for. if it was cheaper than SolusVM, maybe, but it isn't.

    does everyone else write their own?

    free trial zilore monitoring

  • @jmckeag12:

    Why would you store virtual machines in files and not LVM's? I also think you're having some sort of disk corruption because having it totally fail like that wouldn't be on Virtualizor since all it does is command libvirt for KVM.

    Other than that I really don't see how you can have any massive fails if you followed the migration instructions.

  • techhelper1 said: Why would you store virtual machines in files and not LVM's? I also think you're having some sort of disk corruption because having it totally fail like that wouldn't be on Virtualizor since all it does is command libvirt for KVM.

    Even with new Installs. i use it because it has displayed the best results during my tests. How ever there is no simple way back to other storage types without completely reinstalling the nodes in question. but this has nothing to do with the storage type.

  • virtualizor seem to expect LVM. and most info says to use LVM. so far, it's working for me

    free trial zilore monitoring

  • Working Hard with Virtualizor to fix the bugs that are most concerning.

    I'll update this thread and get the title modified if they make me happy again with the Product.

    Thanked by 1Bruce
  • NexHostNexHost Member
    edited August 2015

    I'm happy to announce that Virtualizor have fixed all my present issues. and now that they have done this. I'm happy with their Product again. I have no further bugs with the Product now. I'm now a Registered NOC. and I'm happy to put even more machines online with their panel.

    I now have faith in Virtualizor as they fix any bugs I report to them within 1 hour :)

    Thanked by 1Bruce
  • user error, bugs in virtualizor, or something else? just interested to learn about what happened.

    FYI, I've had all 3 types of problem.

    free trial zilore monitoring

  • @jmckeag12 said:

    they fix any bugs I report to them within 1 hour :)

    about a week quicker than WHMCS!

    how does Virtualizor compare to SolusVM ? has onapp improved things?

    free trial zilore monitoring

  • @Bruce said:
    how does Virtualizor compare to SolusVM ? has onapp improved things?

    OnApp started by reading the documentation. Then realised they need to create some. Currently a few months behind

    Different.

  • NexHostNexHost Member
    edited August 2015

    @Bruce said:
    how does Virtualizor compare to SolusVM ? has onapp improved things?

    I like Virtualizor as they release new content fairly often. and are more then willing to fix the bugs with their software. I have to say that it's hard to find bugs with their software now. couple of improvements could be done. But they are already putting them fixes into the new release. I got at least 4 issues fixed that I was experiencing. and now I'm 110% happy with their product :)

    I'm not sure about SolusVM got fed up using their Product. after waiting more then 2 years for SolusVM 2 to come out. It's a boring product is not updated enough. Virtualizor said to me they are adding support for Proxmox and VMware soon. that is exactly my point as to why I like Virtualizor.

    Support is also way better then SolusVM

    Thanked by 1Bruce
  • Once again another issue occurs similar to the last time. but now on my Ubuntu KVM Nodes. no multivirt.

    The VPS failed to start
    error: Failed to create domain from /etc/libvirt/qemu/v1127.xml
    error: An error occurred, but the cause is unknown

    This error has left customers VMs offline as after they reload them it screws up. and virtualizor support again is unresponsive.

  • NexHostNexHost Member
    edited August 2015

    Look how shit the support is this is all they respond with

     Hi,
    
     There seems to be some issue with vps.
     Can you reboot the main node once ?
    

    What after I reboot the node 3 days ago? That caused one of my clients to go mat at me on a Public Forum and now it has left a VM OFFLINE all Night. and that is all they can respond with.

    Comon everyone switch to Virtualizor and have to reboot your Production nodes every couple of days as that is what the support tell you to do.

  • I have found adding them on Skype, sometimes gets your issue resolved a lot quicker than going back and forwards via the ticket system.

    Have they resolved your issue now?

  • wychwych Member

    Taking a hiatus.

  • Skype: virtualizor

    https://profvps.com - VPS KVM SSD (USA) /w Promo Recurring: 10OFF | Skype: hoang.truong040

  • NexHostNexHost Member
    edited August 2015

    @AshleyUk said:
    I have found adding them on Skype, sometimes gets your issue resolved a lot quicker than going back and forwards via the ticket system.

    Have they resolved your issue now?

    I have them on Skype and it's being looked at but given the reply to reboot the node. is unacceptable. The Node was rebooted the other day to perform a Bios Upgrade. Now that there is a newer kernel version a bug occurs again. resulting in having to downgrade or wait for Virtualizor to fix it. Cannot just reboot the node it will annoy the customers.

     [14:14:43] Dave: panel
     [14:14:48] Dave: hhas left a customers vm offline all night
     [14:15:18] Dave: nope
     [14:15:20] Dave: I submit a ticket
     [14:15:25] Dave: still unanswered still unresolved
     [14:15:30] Dave: and my clients vm is still down
     [14:15:38] Dave: The VPS failed to start
     error: Failed to create domain from /etc/libvirt/qemu/v1127.xml
     error: An error occurred, but the cause is unknown
    
  • jazz1611jazz1611 Member
    edited August 2015

    @jmckeag12 when start that VM, what error show? can you send file xml to me?

    https://profvps.com - VPS KVM SSD (USA) /w Promo Recurring: 10OFF | Skype: hoang.truong040

  • Move to solus. Manually.

    Different.

  • AshleyUkAshleyUk Member
    edited August 2015

    @jmckeag12 said:
    error: An error occurred, but the cause is unknown

    What is outputted when you run :

    virsh define /etc/libvirt/qemu/v1127.xml

    &

    virsh start v1127

    within SSH?

  • @jazz1611 said:
    jmckeag12 when start that VM, what error show? can you send file xml to me?

     Last login: Tue Aug 18 15:43:42 2015 from triband-mum-120.62.**.**.mtnl.net.in
     [email protected]:~# virsh create /etc/libvirt/qemu/v1127.xml
     error: Failed to create domain from /etc/libvirt/qemu/v1127.xml
     error: An error occurred, but the cause is unknown
    
  • NexHostNexHost Member
    edited August 2015

    @AshleyUk said:

     [email protected]:~# virsh define /etc/libvirt/qemu/v1127.xml
     Domain v1127 defined from /etc/libvirt/qemu/v1127.xml
    
     [email protected]:~# virsh start v1127
     error: Domain is already active
    

    And everything is working now. not sure if that helped or Virtualizor have fixed it.

    Ah nvm the VM just auto shuts down after a while.

     2015-08-18 13:53:52.292+0000: 12028: error : printVar:273 : internal error: Buffer too small to print variable 'IP' into
     2015-08-18 13:54:34.504+0000: 12028: warning : qemuDomainObjTaint:1628 : Domain       id=77 name='v1127' uuid=882fe42c-f4f9-46f3-8471-354f3d289b15 is tainted: host-cpu
     2015-08-18 13:56:09.672+0000: 12030: warning : qemuDomainObjTaint:1628 : Domain id=78 name='v1127' uuid=882fe42c-f4f9-46f3-8471-354f3d289b15 is tainted: host-cpu
     2015-08-18 13:57:37.876+0000: 12030: error : printVar:273 : internal error: Buffer too small to print variable 'IP' into
     2015-08-18 13:58:37.088+0000: 12028: warning : qemuDomainObjTaint:1628 : Domain id=80 name='v1127' uuid=882fe42c-f4f9-46f3-8471-354f3d289b15 is tainted: host-cpu
    
  • jazz1611jazz1611 Member
    edited August 2015

    @jmckeag12 maybe same <uuid>xxxx</uuid> with other VM

    https://profvps.com - VPS KVM SSD (USA) /w Promo Recurring: 10OFF | Skype: hoang.truong040

  • Seems for some reason Virtualizor couldn't start the VM, weather its failing to define the QEMU XML file first before trying to start now sure.

    Glad you have got it running now, if Virtualizor GUI ever fails, you can always run the above two commands to get a VM at least up and running / see any errors with the XML config.

  • @jazz1611 said:
    jmckeag12 maybe same <uuid>xxxx</uuid> with other VM

    The bug is to do with IPv6 when it's applied to the VPS. that is what is causing it to fail to start. that is what one of the developers just told me.

  • jazz1611jazz1611 Member
    edited August 2015

    @jmckeag12 Domain id 78 vs 80 was same UUID. Please look it or change different UUID by yourself. maybe it will solved issues.

    https://profvps.com - VPS KVM SSD (USA) /w Promo Recurring: 10OFF | Skype: hoang.truong040

  • @jmckeag12 said:

    it seems there is a issue with your xml syntax in that file as determined by:

    qemuDomainObjTaint:1628 : Domain id=80 name='v1127' uuid=882fe42c-f4f9-46f3-8471-354f3d289b15 is tainted: host-cpu

    I would recommend to go manually into that file with nano or your favorite command line editor and check that out as I had issues like that, that I went into the xml file manually and fixed the issue and didn't have a problem since.

    CubeData FraudRecord Module: https://cubedata.net/fraudrecord OpenNebula module: https://cubedata.net/opennebula now for blesta & whmcs

  • NexHostNexHost Member
    edited August 2015

    @jazz1611 said:
    jmckeag12 Domain id 78 vs 80 was same UUID. Please look it or change different UUID by yourself. maybe it will solved issues.

    There is no domain ID 78 or 80. that seems to be entry logs by the looks of it. and it goes up by number or something.

    Domain ID is 1045 with the vm id being v1127

    The issue is resolved anyway. Thank you everyone on LET for your assistance. :)

    Thanked by 1AshleyUk
  • what problems are you having with IPv6 ? I'm having some fun with assigning subnets.

    a /128 works, but a /112 doesn't unless I also have a /128. waiting for solution from virtualizor

    free trial zilore monitoring

  • NexHostNexHost Member
    edited August 2015

    @Bruce said:
    what problems are you having with IPv6 ? I'm having some fun with assigning subnets.

    a /128 works, but a /112 doesn't unless I also have a /128. waiting for solution from virtualizor

    I have the same issue as you are experiencing also. /112 works fine but anything else does not work. allocating from a /48 IPv6 subnet block.

    Does not work: 2604:d180:0009:0577:0000:0000:000d:0c33

    How ever this will work so is there no DHCPv6?

    2604:d180:0009:270c:0000:0000:0000:0020

  • /112 does work? for me, it isn't.

    free trial zilore monitoring

  • NexHostNexHost Member
    edited August 2015

    @Bruce said:
    /112 does work? for me, it isn't.

    The address provided above does not work. if I manually adjust the config on the VPS and use say 2604:d180:0009:270c:0000:0000:0000:0033 it will work here is a example. give me sometime I'll update this post with a example.

     [[email protected] ~]# ping6 ipv6.google.ccom
     unknown host
    
     inet6 addr: 2604:d180:9:577::13d:6ed5/64 Scope:Global
    
     IPV6ADDR=2604:d180:0009:0577:0000:0000:013d:6ed5/64
    

    If I modify the /64 to a /48 IPv6 works fine. I thought I had allocated a single address not a subnet. as it's not possible to route several /64s as I don't handle the routing. Using a Juniper EX3300 been told that it's not good at handling L3 so I have to let my upstream provider route it as a single /48 or a single /64 won't route any more IPv6 subnets unless we grow and purchase more rackspace.

     [[email protected] network-scripts]# ping6 ipv6.google.com
     PING ipv6.google.com(lax02s21-in-x0e.1e100.net) 56 data bytes
     64 bytes from lax02s21-in-x0e.1e100.net: icmp_seq=1 ttl=50 time=11.1 ms
     64 bytes from lax02s21-in-x0e.1e100.net: icmp_seq=2 ttl=50 time=8.30 ms
    
Sign In or Register to comment.